Saving the Railroad Industry TO Death - The Evil of Economic Freight Rate Regulation

[MichaelSol]

Greyhounds once strongly argued that wheat rates had gone down over the past 25 years, and offered every misreading of every study he could lay his hands on, refusing even to read the explicit texts of the studies he cited to which stated that prices had gone up.

He was vociferous, adamant, expending a great deal of time and energy, even though he has no experience at all with the wheat industry or its rate history-- at one point selecting out of seven or so commonly used inflation indexes the only one that had nothing whatsoever to do with either railroads or agriculture to try and "prove" his point. It was all heat and no light. So much for appropriate methodology.

It was interesting then to see the October 6, 2006 GAO study which showed ... wheat rates had gone up, and in fact GAO singled them out as one of the few categories of rail traffic that had in fact gone up over that time.

So, it is interesting now to see that he has a methodology for "stepping back to check the methodology used."

The wheat rates in question (Montana to the PNW) did decline dramatically in constant dollar terms.  See page 14. 

http://rscc.mt.gov/docs/White_Paper_Meeting_10_05.pdf   (The red line is what the rate would be if it had kept up with the consumer price index.)

The GAO report did not say they went up.

Mr. Sol misrepresents (for whatever purpose) the meaning of the 180% R/VC ratio to the point of falsely claiming that any rail customer above that level is "captive". 

And Mr. Sol has presented absolutely no evidence that any rail customer is being cross subsidized.  He's loudly procliamed that such a thing exists, but saying it don't make it so.

I've got better things to do than argue with someone who I feel misrepresents (out of ignorance, misunderstanding, mallice - I don't know why he does it.) things like this. 

Well, somebody is definitely misrepresenting something.

BLET Newsletter, STB to hold public hearing on grain rates, October 11, 2006:

The financial health of the industry has improved substantially as railroads have cut costs and boosted productivity. Moreover, most rates declined as productivity improvements were passed on to shippers. However, one category of rates examined by GAO--grain rates--diverged from the industry trends. According to the GAO preliminary report, the amount of grain traffic with comparatively high markups over variable cost increased notably between 1985 and 2004.

The Board intends to hold a public hearing after GAO releases its final report, ... as a forum for information about ...  grain transportation markets in general.

Letter, 11/8/06,  JayEtta Z. Hecker, Director, Physical Infrastructure Issues, GAO, to Charles D. Nottingham, Chairman, Surface Transportation Board:

"... At the same time, the 1980 act anticipated that some shippers might not have competitive alternatives-commonly referred to as "captive shippers"-and gave the Interstate Commerce Commission (ICC), and later the Surface Transportation Board (STB), the authority to establish a process so that shippers could obtain relief from unreasonably high rates. This process establishes a threshold for rate relief, allowing a rate to be challenged if it produces revenue equal to or greater than 180 percent of the variable cost of transporting a shipment.

"... Among other things, this report describes the significant changes that have taken place in the railroad industry and reports that from 1985 through 2004, rates generally decreased, but nominal grain rates increased 9 percent. We also found that, on some routes, the amount of grain traveling at rates significantly above the threshold for rate relief had increased since 1985."

White Paper & Briefing Paper For Governor's Schweitzer's Meeting With Vice Chairman Doug Buttrey, STB, Whiteside & Associates, October, 2005:

p. 13. The rail rates in the Northern plains have increased 40% faster than the Rail Cost Adjustment Factor including productivity Unadjusted.

p. 14 The Surface Transportation Board developed specifically for U. S. Railroads a Rail Cost Adjustment Factor which is utilized in measuring Rail Rates and has an adjustment for Productivity Gains. The Montana 52 Car PNW Wheat Rates would be 39 cents/bushel lower today than they are if the BNSF had shared the productivity gains adjustments with Montana farm producers. Comparing rail rates to RCAF has far more relevance than to CPI.

General Accountability Office Report, October 2006, FREIGHT RAILROADS: Industry Health Has Improved, but Concerns about Competition and Capacity Should Be Addressed GAO-07-94:

p. 9: Rail rates generally declined between 1985 and 2000 but increased slightly from 2001 through 2004. Likewise, rail rates have declined since 1985 for certain commodity groups and routes despite some increases since 2001, but rates have not declined uniformly, and some commodities are paying significantly higher rates than others. For example, from 1985 through 2004, coal rates declined 35 percent while grain rates increased 9 percent.

p. 13: Grain rates initially declined from 1985 through 1987, but then diverged from the other commodity trends and increased, resulting in a net 9 percent increase by 2004.

 

[Datafever]

The wheat rates in question (Montana to the PNW) did decline dramatically in constant dollar terms.  See page 14. 

http://rscc.mt.gov/docs/White_Paper_Meeting_10_05.pdf   (The red line is what the rate would be if it had kept up with the consumer price index.)

The GAO report did not say they went up.

...

You are correct in saying that the GAO report did not say that wheat rates had gone up.  In fact, the GAO report never separates out wheat from other grains.

Here is what the GAO report does say:

Rail rates generally declined between 1985 and 2000 but increased slightly from 2001 through 2004. Likewise, rail rates have declined since 1985 for certain commodity groups and routes despite some increases since 2001, but rates have not declined uniformly, and some commodities are paying significantly higher rates than others. For example, from 1985 through 2004, coal rates declined 35 percent while grain rates increased 9 percent.

[greyhounds]

Greyhounds once strongly argued that wheat rates had gone down over the past 25 years, and offered every misreading of every study he could lay his hands on, refusing even to read the explicit texts of the studies he cited to which stated that prices had gone up.

He was vociferous, adamant, expending a great deal of time and energy, even though he has no experience at all with the wheat industry or its rate history-- at one point selecting out of seven or so commonly used inflation indexes the only one that had nothing whatsoever to do with either railroads or agriculture to try and "prove" his point. It was all heat and no light. So much for appropriate methodology.

It was interesting then to see the October 6, 2006 GAO study which showed ... wheat rates had gone up, and in fact GAO singled them out as one of the few categories of rail traffic that had in fact gone up over that time.

So, it is interesting now to see that he has a methodology for "stepping back to check the methodology used."

The wheat rates in question (Montana to the PNW) did decline dramatically in constant dollar terms.  See page 14. 

http://rscc.mt.gov/docs/White_Paper_Meeting_10_05.pdf   (The red line is what the rate would be if it had kept up with the consumer price index.)

The GAO report did not say they went up.

Mr. Sol misrepresents (for whatever purpose) the meaning of the 180% R/VC ratio to the point of falsely claiming that any rail customer above that level is "captive". 

And Mr. Sol has presented absolutely no evidence that any rail customer is being cross subsidized.  He's loudly procliamed that such a thing exists, but saying it don't make it so.

I've got better things to do than argue with someone who I feel misrepresents (out of ignorance, misunderstanding, mallice - I don't know why he does it.) things like this. 

[Datafever]

My understanding is that additional volume approaching capacity limits is a good thing.  ...   Underutilized capacity is not normally a good thing.

This is where, in my view, railroading separates from the manufacturing model.  In a broad view, I disagree with both of these statements, but oh my gosh ... that's a thread to itself ...

Well, it isn't like this thread is anywhere near its original discussion, nor have I seen anyone else trying to veer it back to the issue of regulation, so please continue.  Expound on the reasons behind your enigmatic statement.  I'll keep in mind that you are taking the broad view, and not specific micro-examples. 

EDIT: Since you edited your post, I'm editing mine.

I would define capacity as that point at which inefficiencies enter the system.  But bear in mind that my statement was "approaching capacity", not "at capacity".  

[MichaelSol]

My understanding is that additional volume approaching capacity limits is a good thing.  ...   Underutilized capacity is not normally a good thing.

This is where, in my view, railroading separates from the manufacturing model.  Railroading has a range of capacity that is economically efficient. Above or below it, and it is inefficient. These guys that argue that railroads should operate "at capacity" have never run the numbers of just how expensive it is to run a railroad "at capacity." In a broad view, I disagree with both of these statements, but oh my gosh ... that's a thread to itself ...

[Datafever]

Mr. Sol, I guess that I am just going to have to pass on this one until there are some things that I understand a little better.

I think that I basically agree with you, but it would appear that the terminology that I have used is inconsistent with "proper" terminology.

My understanding is that additional volume approaching capacity limits is a good thing.  Reaching or exceeding capacity limits is a bad thing.  Underutilized capacity is not normally a good thing.

For any distributed network that is not perfectly balanced (which would include practically all real-world networks), there will always be routes on that network that will become bottlenecks as network traffic increases.  Increasing capacity on those routes will cause other routes to become bottlenecks as traffic continues to increase.  If a network is well distributed, traffic will be more balanced.

[MichaelSol]

  But how does one determine if cross-subsidization is taking place then? 

A CVP analysis -- cost, volume, price -- is how a manufacturing firm analyzes a combination of rates at given volumes for its product lines or services.

I posted the following a couple of pages ago, and I wouldn't know what to add to it.

"Pretty good summary. At that point, both volume and capacity become key factors. With excess capacity, anytime the company can sell the product at any percentage over the VC, the company earns revenue to cover fixed costs.

"Enough volume, and the product or commodity can in fact generate a profit in excess of the total fully distributed costs (FC + VC(x)) of the service for that product or commodity. Indeed, as cross-subsidized traffic adds volume, the fixed cost allocation per unit for all traffic declines, most of the time, and other traffic becomes more profitable relative to their fully allocated costs as a result. Cross subsidization does not exist in a vacuum and has both positive and negative potential impacts on the profitability of other services as well as the company as a whole.

"However, until the service in question covers its own variable and fixed costs, it is cross-subsidized. After that point it is a genuine profit center.

"At close to capacity, cross-subsidized products or services suddenly develop sharp edges. The company can't just keep adding volume. The capacity constraint may prevent development or acquisition of more profitable traffic. And, this is more true for railroads than just about any other industry -- as capacity limits are approached, costs of operation -- variable costs -- begin to climb. Cross-subsidized traffic that may have been contributing to  fixed costs suddenly gets swallowed up by rising variable costs.

"The rate of climb in operating costs for Class I's over the past 4 years is unprecedented in a low inflation environment. Slower trains, dead on hours, all of the indicators of capacity limits all increase operating costs. This is why the average R/VC has slid over the past six years so that more traffic now travels at rates under 180% R/VC than six years ago -- and why more traffic is now forced to travel at rates over 300% R/VC to make up for it."

The idea of allocating fixed costs is a management tool. It is commonly called absorption cost accounting. But, it is stand-alone -- used to measure the quality of a given product or service in terms of profitability -- but which becomes critical where capacity is a constraint. Can't use it in CVP analysis.

 

[Datafever]

Okay, we seem to have come full circle on this discussion.  As I recall, the expression of fixed costs as a percentage of variable costs was first brought up by you as a means to show that a particular revenue level was profitable.

For a company as a whole, there is a historic relationship of FC to VC. FC run about 35% of VC. Taking it further than that, and attempting to say that 37% FC/VC, or 28% FC/VC, or some other ratio is taking the historical relationship to a place that it cannot go. My point was, what's the point?

We don't know what a high FC/VC rato means, or a low one for that matter. So, while there is a historical record that stays fairly constant of fixed costs to variable costs, I did not and do not see the analytical purpose of attempting an analysis based on a FC/VC ratio since for analytical purposes the analysis doesn't report anything useful.

An historical observation did not translate into an analytical tool. 

Okay, point taken.  But how does one determine if cross-subsidization is taking place then? 

[MichaelSol]

Okay, we seem to have come full circle on this discussion.  As I recall, the expression of fixed costs as a percentage of variable costs was first brought up by you as a means to show that a particular revenue level was profitable.

For a company as a whole, there is a historic relationship of FC to VC. FC run about 35% of VC. Taking it further than that, and attempting to say that 37% FC/VC, or 28% FC/VC, or some other ratio is taking the historical relationship to a place that it cannot go. My point was, what's the point?

We don't know what a high FC/VC rato means, or a low one for that matter. So, while there is a historical record that stays fairly constant of fixed costs to variable costs, I did not and do not see the analytical purpose of attempting an analysis based on a FC/VC ratio since for analytical purposes the analysis doesn't report anything useful.

An historical observation did not translate into an analytical tool.

 

 

[Datafever]

"But in trying to determine if cross-subsidization is occurring or if the railroad is making unreasonable profits, the ability to determine the FC/VC ratio becomes necessary.  Depending upon how fixed costs are "allocated" (poor choice of words on my part - I apologize), it may very well be possible that a shipper receiving a rate of 120% R/VC is not being cross-subsidized, or that a shipper paying 160% R/VC is being cross-subsidized. "

An "FC/VC ratio" is outside my experience or education in marginal cost pricing theory. Never heard of it. I think that has to do with the fact that marginal cost pricing theory as used by the Staggers Act creates a mathematical threshold that includes an absolute, which is the variable cost. The railroad receives 100% of the variable cost.  Always. No matter what it is.

On a percentage scale, there are only two mathematically exclusive circumstances, 0% and 100% -- they are exclusionary of alternatives existing. Just like dividing by zero creates a mathematical impossibility, trying to create a "percentage" analysis out of a 100% cost formula yields complete gibberish.  

I was so puzzled  by the notion of an FC/VC ratio, I created a model to see what in the heck people are talking about by a "FC/VC ratio."

...

If a commodity moves at $2800 a carload, and the company has $300  in fixed costs on a fully distributed basis, and variable costs are $2000 -- the commodity will have a 140% R/VC ratio, and earn 22% net on a fully distributed basis. The FC/VC ratio in that instance is 15%.

Suppose the commodity moves on a shorter haul -- lower VC, lower revenue, right? Say $1,000 VC, and $1600 revenue. Fixed cost allocation has to stay the same at $300, but varies in relation to the variable cost (unless the company allocates on the basis of ton-miles). The FC/VC ratio has doubled to 30%! Gasp. What does that mean? It means in that case that the R/VC increases to 160%, and profitability increases to 23%.

Well, what does that mean? A high FC/VC ratio is bad? Good?  

To me at least, an FC/VC ratio is meaningless.

Okay, we seem to have come full circle on this discussion.  As I recall, the expression of fixed costs as a percentage of variable costs was first brought up by you as a means to show that a particular revenue level was profitable.

Let's go back to widget building.  If a company designs and builds multiple types of widgets, it will no doubt be able to define the "per unit" manufacturing cost of each widget type.  But unless it has some way to identify the "fixed costs" associated with a particular widget type, how can it ever determine if that widget is "profitable" for the company or not?  Or is it sufficient to decide that if all the fixed costs are being covered by the revenue generated by the sales of all the widgets, then profitability for a particular LOB is irrelevant? 

[MichaelSol]

Fixed cost expressed as a percentage of total cost vary (don't get confused) with volume.  So the 35% figure will, in fact, "vary all over the place" with volume on a particular line segment.  This figure is a percentage of total cost.  If the percentage of cost that are variable goes up, the percentage of cost that are fixed must go down.

Fixed costs are "operating" costs. They arise from centralized shop and mechanical facilities, administrative overhead, annualized expenses such as financing and taxes associated with physical plant.  Most of these are not directly associated with the "line" at all. Most states even charge ad valorem property taxes on rail plant -- not on the valuation of a given "line" in the state. Can't even get out from under a general mortgage that may be paying off a "line" --  sure enough, the general mortgage survives whether the line is there or it is abandoned -- whether it has all the traffic or none of it. That's why certain costs are considered overhead.

What you are saying is that if a company were to be completely arbitrary and say that overhead costs can be allocated to a geographical entity such as a line, and then divert traffic to that line -- say double it -- the company can indeed show that overhead costs per unit volume would decline, by 50% in that case. In that instance, and here is the conundrum of your proposition, the fixed cost per unit volume varies "all over the place" only when you allocate it to the actual traffic on the line -- but you say you are not doing that and yet that is exactly what you are doing!

But, if the traffic on the system had legitimately doubled -- overhead will increase. Historically, overhead pretty much tracks overall variable costs as a given percentage. The only way fixed charges could vary all over the place is by cooking the overhead books, isolate an example, artificially redefine overhead, and voila' ... 

"But in trying to determine if cross-subsidization is occurring or if the railroad is making unreasonable profits, the ability to determine the FC/VC ratio becomes necessary.  Depending upon how fixed costs are "allocated" (poor choice of words on my part - I apologize), it may very well be possible that a shipper receiving a rate of 120% R/VC is not being cross-subsidized, or that a shipper paying 160% R/VC is being cross-subsidized. "

An "FC/VC ratio" is outside my experience or education in marginal cost pricing theory. Never heard of it. I think that has to do with the fact that marginal cost pricing theory as used by the Staggers Act creates a mathematical threshold that includes an absolute, which is the variable cost. The railroad receives 100% of the variable cost.  Always. No matter what it is.

On a percentage scale, there are only two mathematically exclusive circumstances, 0% and 100% -- they are exclusionary of alternatives existing. Just like dividing by zero creates a mathematical impossibility, trying to create a "percentage" analysis out of a 100% cost formula yields complete gibberish.  

I was so puzzled  by the notion of an FC/VC ratio, I created a model to see what in the heck people are talking about by a "FC/VC ratio."

Looking at two extreme examples on the model.  A commodity travels at $2800 per carload. If the fixed cost is $300 per carload, and the variable costs vary between $100 and $3500, the "FC/VC ratio" varies between 300% -- at which the commodity is earning a 2800% R/VC ratio and a 600% net profit and an FC/VC ratio of 10%, at which the R/VC is 93% and net profit is -15%.

Another model variant, the variable cost is $1,000. What happens if fixed cost "varies" from $100 per carload to  $3,000 per carload? The results when the FC/VC ratio is 10% is a 280% R/VC and a net profit of 155%.

In other words, the FC/VC ratio tells the company absolutely nothing, nothing at all.  

The reason is that the VC is an undefined number. It is 100% of some reality which does, in fact, vary. Fixed costs are not directly related to the variable costs of the move. A "percentage" analysis by means of an FC/VC ratio is just gibberish -- if it means anything at all in a particular instance, it is an accident.

If a commodity moves at $2800 a carload, and the company has $300  in fixed costs on a fully distributed basis, and variable costs are $2000 -- the commodity will have a 140% R/VC ratio, and earn 22% net on a fully distributed basis. The FC/VC ratio in that instance is 15%.

Suppose the commodity moves on a shorter haul -- lower VC, lower revenue, right? Say $1,000 VC, and $1600 revenue. Fixed cost allocation has to stay the same at $300, but varies in relation to the variable cost (unless the company allocates on the basis of ton-miles). The FC/VC ratio has doubled to 30%! Gasp. What does that mean? It means in that case that the R/VC increases to 160%, and profitability increases to 23%.

Well, what does that mean? A high FC/VC ratio is bad? Good?  

To me at least, an FC/VC ratio is meaningless.

 

 

[Datafever]

  I would think that if a shipper doubles the amount of material shipped from one location to another, then the only cost that has changed is the VC.  In fact, the VC would have almost doubled, no? 

This is only a technical point; I think it is clear what you mean.

"Variable Cost" is conventionally defined as the cost per unit. That variable cost exists on a curve that is variable with reference to production run and economic unit size. You could, perhaps, have a "Total Variable Cost" but it would instantly change with the addition or loss of one unit of production.

The Cost Equation uses VC as the single unit cost; TC=FC+VC(x). 

Mr. Sol, I have been pondering what you said all day, trying to come to grips with it.

Part of me understands what you are saying, but part of me is having problems with trying to adjust it to the specifics of railroading.

So, let me fall back to what I know better than nothing.  In reading the STB reports, it appears fairly clear to me that a shipper's total volume is used when trying to determine R/VC.  As volume is irrelevant to that equation (increased volume = increased revenue = increased variable costs), that is not an issue.

But in trying to determine if cross-subsidization is occurring or if the railroad is making unreasonable profits, the ability to determine the FC/VC ratio becomes necessary.  Depending upon how fixed costs are "allocated" (poor choice of words on my part - I apologize), it may very well be possible that a shipper receiving a rate of 120% R/VC is not being cross-subsidized, or that a shipper paying 160% R/VC is being cross-subsidized. 

[Datafever]

Thank you, arbfbe.  You are correct in pointing out that doubling the "volume" will not always cause an increase in variable costs, and I expect that a doubling of VC would probably be the upper limit.

Your post caused me to reflect on other situations too.  For instance, if a doubling of volume caused capacity problems such as congestion that required double-tracking buildout or additional yards, then the fixed costs could also increase.  In such a case, FC could potentially increase to a greater extent than VC, causing the FC/VC ratio to increase instead of decrease.

In a nutshell, increased volume can decrease the FC/VC ratio if the capacity is available to handle the additional volume, but could increase the FC/VC ratio in certain other cases.

As someone else said, it is always possible to build scenarios where the "general" case falls apart.  IMO, it doesn't mean that such scenarios should not be investigated.  In some cases, the special cases may help identify a broader range of dynamics for establishing the general case.

When all is said and done, I can understand why the FC/VC ratio tends to stay fairly steady for a large (Class 1) railroad. 

[arbfbe]

Mr. Fever,

Say the shipper is moving ten cars per day of product.  Now he increases that to twenty cars per day.  He has indeed doubled the quantity shipped but since neither amout is enough to require a dedicated move the variable costs have not doubled.  Crew costs will remain the same so long as all twenty cars move in the same train, fuel costs will rise incrementally and there may be some additional wear and tear on the infrastructure.  Most additional variable costs will be barely measurable.

Now if the increase is from 100 cars per day to 200 cars per day then variable costs will be very nearly doubled.  Twice as many crews, twice as many locomotives (probably anyway, unless the power from the first train is doubled back with the mtys to make the second trip), twice as much fuel, twice as much dispatching time and so on and so on.  

   

[MichaelSol]

  I would think that if a shipper doubles the amount of material shipped from one location to another, then the only cost that has changed is the VC.  In fact, the VC would have almost doubled, no? 

This is only a technical point; I think it is clear what you mean.

"Variable Cost" is conventionally defined as the cost per unit. That variable cost exists on a curve that is variable with reference to production run and economic unit size. You could, perhaps, have a "Total Variable Cost" but it would instantly change with the addition or loss of one unit of production.

The Cost Equation uses VC as the single unit cost; TC=FC+VC(x). 

[MichaelSol]

No, you can properly and accurately "assign" fixed cost to a line segment such as a branch line.  ...

It's fairly simple to identify such line specific fixed cost.  

greyhounds:

10-15-2006, "STB to Hold Hearings on Grain Shipments":

"What Mr. Sol continualy fails to understand is that you can not allocate fixed cost."

What I said a few post back was that while you can not allocate fixed cost to business moving over a line, you can identify fixed cost that will cease to exist if the line is abandoned.  The cost must be covered by the traffic on the line, but can not be allocated amongst the business on the line.

There is no contradiction between my two statements.

Well, there is. One explicitly contradicts the other. Line allocation is, in fact, "an allocation" of the type you rejected when the argument was convenient to do so on another occasion.

Operating costs are generated by traffic. Not by "a line."  Fixed costs are a version of operating cost. Administrative overhead in Chicago is paid for by all traffic, and will not likely increase or decrease one iota as the result of a line abandonment.

Fixed costs are borne by the company -- that is, the total of traffic or service.

Now, abandonment costs. Is that an operating consideration? No. In explicitly the same fashion that investment decision analysis through internal rate of return methodology requires "allocation,"  neither abandonment nor investment are operating cost considerations.

You can't allocate anything that is not an operating cost to operating costs, and we in fact are talking about operations, not investment or abandonment. Those are capital or extraordinary charges and not even remotely classified as operating costs, notwithstanding an investment, regulatory or SEC reporting requirement to identify the charge or the write-down as the case may be.

For general operating considerations, fully distributed cost allocation goes to the traffic or market share, not the building (line) housing the operation or a portion of it. 

See: 

"Profitability Analysis by Market Segments", Leland L. Beik; Stephen L. Buzby, Journal of Marketing,  37:3, Jul., 1973.

"The Problem of Fixed Charges", Donald L. Raun, The Accounting Review, 26:3, July, 1951, January, 1961.

"Overhead Costs and Income Measurement", William Ferrara, The Accounting Review, 36:1.

"Separation of Fixed and Variable Costs", Julius Wiener, The Accounting Review, 35:4, October, 1960.

"Bases for Allocating Distribution Costs", Robert H.  Watson, Journal of Marketing, 16:1, July, 1951.

"Profit Maximizing Cost Allocation Using Cost-based Pricing", Teresa Pavia, Management Science, 41:6, June, 1995.

"Cost Allocation in Transportation Systems", Robert C.  Anderson, Armin Claus, Southern Economic Journal, 43:1, July, 1976;

 

 

[Datafever]

I have one of those stuck-on-stupid types of questions.

Could someone explain to me how studies (including the GAO report) determine R/VC for a particular shipper?  I would think that it can't be completely straightforward, or the rate relief process through the STB would not take 3+ years.

[greyhounds]

No, you can properly and accurately "assign" fixed cost to a line segment such as a branch line.  ...

It's fairly simple to identify such line specific fixed cost.  

greyhounds:

10-15-2006, "STB to Hold Hearings on Grain Shipments":

"What Mr. Sol continualy fails to understand is that you can not allocate fixed cost."

 

What I said a few post back was that while you can not allocate fixed cost to business moving over a line, you can identify fixed cost that will cease to exist if the line is abandoned.  The cost must be covered by the traffic on the line, but can not be allocated amongst the business on the line.

There is no contradiction between my two statements.

[MichaelSol]

No, you can properly and accurately "assign" fixed cost to a line segment such as a branch line.  ...

It's fairly simple to identify such line specific fixed cost.  

greyhounds:

10-15-2006, "STB to Hold Hearings on Grain Shipments":

"What Mr. Sol continualy fails to understand is that you can not allocate fixed cost."

 

[Datafever]

Well, the fixed costs don't vary all over the place, per se, but the variable costs certainly can.  I would think that if a shipper doubles the amount of material shipped from one location to another, then the only cost that has changed is the VC.  In fact, the VC would have almost doubled, no? 

Well, you're right.  The variable cost would go up - I don't know about doubling.

But what we were talking about with the 35% figure was the percentage of cost that were fixed. 

If the variable cost go up (double?) and the fixed cost remain constant (which they must do or they are not "fixed" cost) then the percentage of fixed cost as a component of total cost must go down.  Which will cause the 35% figure to change.

Fixed cost expressed as a percentage of total cost vary (don't get confused) with volume.  So the 35% figure will, in fact, "vary all over the place" with volume on a particular line segment.  This figure is a percentage of total cost.  If the percentage of cost that are variable goes up, the percentage of cost that are fixed must go down.

Exactly - that's the point that I was trying to make.  In your original post, you didn't claim that the fixed costs would vary, only that the 35% FC to VC would vary.  If VC doubles, then the FC/VC ratio would halve. 

[greyhounds]

Well, the fixed costs don't vary all over the place, per se, but the variable costs certainly can.  I would think that if a shipper doubles the amount of material shipped from one location to another, then the only cost that has changed is the VC.  In fact, the VC would have almost doubled, no? 

Well, you're right.  The variable cost would go up - I don't know about doubling.

But what we were talking about with the 35% figure was the percentage of cost that were fixed. 

If the variable cost go up (double?) and the fixed cost remain constant (which they must do or they are not "fixed" cost) then the percentage of fixed cost as a component of total cost must go down.  Which will cause the 35% figure to change.

Fixed cost expressed as a percentage of total cost vary (don't get confused) with volume.  So the 35% figure will, in fact, "vary all over the place" with volume on a particular line segment.  This figure is a percentage of total cost.  If the percentage of cost that are variable goes up, the percentage of cost that are fixed must go down.

Or, as I once learned in an accounting class; "Variable cost are fixed and fixed cost are variable."  That means that variable cost remain constant per unit of production while fixed cost expressed per unit of productiion fluctuate (vary all over the place) with volume. 

[Datafever]

The 35% figure will vary all over the place.

How much?

 

  In general. A high volume line will have a lower fixed to variable ratio than a low volume line.

 

"Overhead [fixed] costs  ... are typically incurred in the production of all of the services provided by a firm." Ronald Braeutigam, "An Analysis of Fully Distributed Cost Pricing in Regulated Industries,"  The Bell Journal of Economics, 11:1, Spring, 1980, pp. 182-196, 184. 

Fixed costs are company costs, not line costs. 

By definition, they can't "vary all over the place."

 

Well, the fixed costs don't vary all over the place, per se, but the variable costs certainly can.  I would think that if a shipper doubles the amount of material shipped from one location to another, then the only cost that has changed is the VC.  In fact, the VC would have almost doubled, no? 

[greyhounds]

The real problem is allocating the revenue to a specific line segment.  How much revenue from a through move that starts on a branch and ends on a main line do you allocate to the branch?  There's no good answer to that one. 

In the 1970s at the C&NW we just went with the ICC regs.  They told us to allocate 50% of our division of the revenue to the branchline under study. If an off branch movement cost $1,000 and we got 25% of the revenue we would take 50% of our $250 and apply that $125 to the revenue account on the branch.

We went with the ICC formula to avoid hours of pointless testimony about how to allocate the revenue and the costs.  If someone objected to the formula we just said we were only following the ICC's rules.

We did the same at the ICG.  There was no choice.  It was obvious that the 50% figure was pulled out of the air, but that was the ICC regulation and the ICC had to approve the abandonment.

The regulators had all kinds of detailed rules for an abandonment.  Then they pulled this 50% figure out of the air.   I thought it was unreal.  It was very real.  That's the best they could come up with.

 I remember one of the rules was that we had to notify the top 10 customers on a line that we were going to abandon the line.   But many of the lines didn't have 10 customers.   So I went and ask a lawyer what to do in that situatiion.  I never did get an answer. 

[bobwilcox]

The real problem is allocating the revenue to a specific line segment.  How much revenue from a through move that starts on a branch and ends on a main line do you allocate to the branch?  There's no good answer to that one. 

In the 1970s at the C&NW we just went with the ICC regs.  They told us to allocate 50% of our division of the revenue to the branchline under study. If an off branch movement cost $1,000 and we got 25% of the revenue we would take 50% of our $250 and apply that $125 to the revenue account on the branch.

We went with the ICC formula to avoid hours of pointless testimony about how to allocate the revenue and the costs.  If someone objected to the formula we just said we were only following the ICC's rules.

[greyhounds]

The 35% figure will vary all over the place.

How much?

 

  In general. A high volume line will have a lower fixed to variable ratio than a low volume line.

 

"Overhead [fixed] costs  ... are typically incurred in the production of all of the services provided by a firm." Ronald Braeutigam, "An Analysis of Fully Distributed Cost Pricing in Regulated Industries,"  The Bell Journal of Economics, 11:1, Spring, 1980, pp. 182-196, 184. 

Fixed costs are company costs, not line costs. 

By definition, they can't "vary all over the place."

No, you can properly and accurately "assign" fixed cost to a line segment such as a branch line.  You can't allocate those cost to various traffic segments moving over the line, but they can be line specific.

There certainly are cost elements for any line that are 'fixed' in that they do not vary with the amount of traffic moved on the line.  Examples are taxes and ownership cost on the property.  Since these don't change with traffic volume they are 'fixed' and can not be allocated to any particular business on the line.  If the business goes away, they don't change.

However, if the rail line goes away, so do those expenses.  It's fairly simple to identify such line specific fixed cost.   A goal of accounting is to match revenues with expenses.  Sometimes you can do it well, sometimes there is no match to be found.  But the line specific fixed charges can be matched to the line and they're a big consideration in branch line abandonments.

The real problem is allocating the revenue to a specific line segment.  How much revenue from a through move that starts on a branch and ends on a main line do you allocate to the branch?  There's no good answer to that one. 

There are common corporate expenses, such as heating the HQ building, that can not be assigned to any particular piece of the business.  But big elements of fixed expense on a particular line can be properly assigned to the line - but not allocated amongst the traffic on the line.

 

 

[MichaelSol]

Almost all coal moves under contract.  Coal accounts for nearly 50 percent of tonnage but only 23 percent of revenue for Class I railroads.  Of the remaining traffic by tonnage, this means 40% moves under contract and 60% moves by tariff. Of the 40% that moves under contract, any bets as to who gets the genuine competitive advantage -- competitive or non-competitive shippers? Who has the bargaining power?

If what you are saying is correct (regarding coal tonnages and revenue), then the other 20% of contract tonnage accounts for 48% of Class 1 revenues.  That doesn't sound like much of a price break to me.

Well, Hunt, UPS -- there are some contracts out there that are not of the coal variety. 

The difference is between when the contracts were made, and now. Bob's post goes to this situation.

Railroads were trying to lock shippers into contracts to avoid the free fall in rates that was occuring. Mainly, it locked railroads into ultimately non-compensatory prices. Those contracts have been expiring and are expiring. 

The rationale behind the contracts was railroad driven -- a miscalculation ultimately on their part, but not really useful in this discussion as it obscures, and does not illuminate, the captive shipper/competitive shipper problem.

Depends also on the "when" of the contract. Current coal contracts renegotiated this year are at two to three times the former price. My coal revenue/tonnage numbers were from 2002 or 2001.  Captive  shippers who need contracts -- coal is one -- are currently renegotiating in a substantially different environment than 15-20 years ago.

Railroads contract to avoid price drops. Shippers contract to avoid price increases. Depending on which way the price actually goes defines who gets the benefit. Contracts are a "hedge", and do not necessarily reflect markets.

 

 

 

[Datafever]

Michael - I thought grain shippers had to use tariffs but this article talks about a contract.  Will the terms of this contract be made public?

I would expect the ratio of contract to tariff movements is shifting to tariffs.  Twenty years ago a lot of small movements were put in contracts for a variety of reasons. One year at the SP we had a contest where the salesmen who negotiated the most contracts got a very nice one week vacation in Maui for they and their significant other.  Many times a 100 car package got split into four 25 car contracts. Oh well, it was a great week at the Hyatt.

There has been a push on the get these small moves out of contracts and into tariffs to reduce the back office burden.

Speaking of which, shippers are lamenting the move from contracts to published rates, based on the argument that published rates reduce competitiveness. 

[bobwilcox]

Michael - I thought grain shippers had to use tariffs but this article talks about a contract.  Will the terms of this contract be made public?

I would expect the ratio of contract to tariff movements is shifting to tariffs.  Twenty years ago a lot of small movements were put in contracts for a variety of reasons. One year at the SP we had a contest where the salesmen who negotiated the most contracts got a very nice one week vacation in Maui for they and their significant other.  Many times a 100 car package got split into four 25 car contracts. Oh well, it was a great week at the Hyatt.

There has been a push on the get these small moves out of contracts and into tariffs to reduce the back office burden.

[Datafever]

Almost all coal moves under contract.  Coal accounts for nearly 50 percent of tonnage but only 23 percent of revenue for Class I railroads.  Of the remaining traffic by tonnage, this means 40% moves under contract and 60% moves by tariff. Of the 40% that moves under contract, any bets as to who gets the genuine competitive advantage -- competitive or non-competitive shippers? Who has the bargaining power?

If what you are saying is correct (regarding coal tonnages and revenue), then the other 20% of contract tonnage accounts for 48% of Class 1 revenues.  That doesn't sound like much of a price break to me.